September 4, 2023 • 46 mins
Why do you still feel the waves after getting off a boat? Why does the wall seem to come at you faster after you step off the treadmill? Why do the rocks seem to move upward after you stare at a waterfall? Why did people in the 1980s think their book pages had some red color in them… but no one thought that before or after the 80s? And what does any of this have to do with drugs, heartbreak, yellow sunglasses, or Aristotle watching a horse stuck in a river? Join Eagleman to understand how the brain constantly readjusts its circuitry to best read the world, and what it means for our (sometimes strange) perceptions of what's out there.
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Speaker 1 (00:04):
Why do you still feel the waves after you get
off a boat and when you get off the treadmill
at the gym. Why does everything seem to be flowing
past you faster? What does heartbreak have in common with
drug withdrawal? And why after you stare at a waterfall
for a while, do the rocks on the side seem
to be crawling upward? Why did people in the nineteen
(00:27):
eighties think their book pages had some color red in them,
but no one thought that before or after the eighties?
And what does any of this have to do with
the great philosopher Aristotle watching a horse stuck in the river?
Welcome to Inner Cosmos with me David Eagleman. I'm a
(00:49):
neuroscientist and author at Stanford and in these episodes we
sail deeply into our three pound universe to understand why
and how our lives look the way they do. Today's
(01:09):
episode is about the way the brain readjusts its circuitry
on really fast timescales, and why your brain is constantly
doing this. In the nineteen eighties, tens of thousands of
people began to notice something really weird when they looked
at a floppy disk envelope with the black and white
(01:31):
IBM logo on the front, the letters IBM seemed to
have a red tint, And the same thing happened when
people looked at pages in a book, the text seemed
like it was shaded red. Now, if that's not weird enough,
check this out. This only happened in the nineteen eighties.
People didn't perceive a red tint before or after this decade.
(01:54):
So what was changing about brains during that window. I'll
tell you the answer. But to understand this, we're first
going to step back even further in history, twenty four
hundred years further to Aristotle. Now, Aristotle, as you know,
is a philosopher and a polymath whose intellectual interests were boundless.
(02:18):
He wrote about essentially every subject you could imagine. But
we're coming to Aristotle not because of something he wrote,
but instead because he was insightful enough to pause and
notice when something didn't make sense. What happened is that
someone was trying to cross a fast flowing river on
his horse, and the horse got stuck halfway in the river.
(02:41):
So Aristotle, along with some other people, ended up watching
this rescue operation. The way you might have a bunch
of people looking at a car crash nowadays. So you
have all these Greeks all standing there in their togas,
talking and chatting and watching the horse in the fast
flowing river. Now most of the people were presumably just
talking about stuff that made no difference to history. But
(03:03):
Aristotle stood at the side of the river intently watching
this horse. Now Here was the key about Aristotle, a
general characteristic that made him so famous to history. He
was a good observer. He paid attention. He noticed when
things were unusual, and when he found something, he would
(03:25):
sink his teeth into the problem and wouldn't let go.
And what he noticed here was something that possibly the
others did too, but they ignored it, and they didn't
think any further on it. What he noticed is that
once the horse got unstuck and the man made it
to the shore, and Aristotle looked away, he noticed that
everything on the land, the rocks, the huts, the trees,
(03:47):
everything seemed to be drifting, and they all seemed to
be moving in the opposite direction of the river. Now
you may have seen something like this too. The easiest
way to experience Aristotle's confusion and delight is to stare
at a waterfall and after you keep your eyes locked
on it for a while, look over to the rocks
(04:10):
on the side of the waterfall, and the rocks appear
to be moving upward. Now, this illusion that he first
made note of twenty four hundred years ago has come
to be known as the motion after effect. Why does
it happen. The activity of particular neurons in your visual
cortex represents downward motion, and the activity of other neurons
(04:35):
represents upward motion. And they're always locked in battle, and
most of the time the competition is evenly pitched, and
they evenly inhibit each other, and as a result, the
world appears to you to be moving neither up nor down. So,
given this, a popular explanation for the motion after effect
(04:56):
is fatigue. By staring at the downward motion of the waterfall,
you burn up a good deal of energy in your
downward coating neurons, and now their vigor is temporarily depleted.
So the battle tips in favor of the upward encoding neurons,
and as a result of this unbalanced activity, you perceive
(05:19):
that the rocks are climbing upward. You have this net
movement upward. Now, this fatigue hypothesis is very appealing in
its simplicity, but it's wrong. Why because it can't explain
some critical facts about the illusion. So imagine you watch
the downward waterfall for a while and then you close
(05:41):
your eyes tightly, say, for three hours you close your eyes.
When you reopen your eyes, you'll see the rocks crawling upward.
And what that tells us is it's not about a
temporary energy depletion in neurons. There's something deeper going on here.
And here's what that deeper thing is is the illusion
(06:01):
comes about not because of passive fatigue, but instead because
of an active recalibration in your brain. Your visual system
is exposed to continuous downward motion, and after a while
it comes to assume this is the new normal. At first,
downward motion is dramatic information to your brain, but after
(06:25):
a while of staring, you receive no new information from
downwardness here. So, as far as your brain is concerned,
this is the new reality, a world that flows more
down than up. So your visual system carefully rebalances its
expectations to mirror the world to expect more downward than
(06:47):
upward activity. Now, when you look away from the waterfall
and toward the cliff side. This recalibrated set point becomes
obvious because now rocks and trees are flowing upward toward
the sky. Your set point has shifted. In other words,
what counts as standing still has shifted. Why well, your
(07:10):
brain always wants to set up a ground truth so
that it can be better at detecting change. In this case,
when your visual field is filled with the site of
the waterfall, your brain is working to subtract off all
that downward motion. All that downward motion is no longer
(07:30):
informative in the way that it was, and so the
circuitry it adjusts itself so that it can be maximumly
sensitive to new information. Now, if you watch for it
the way Aristotle would, you'll see that this kind of
recalibration happens to you all the time. Think about when
you're on a small boat on the ocean. You are
(07:51):
rocking with the waves for a while, and then when
you get off the boat, the land seems to be
rocking for a while. It feels as though you're still
on the water. Now you may have thought about it
this way, Hey, my body thinks it's still on the water.
What you're actually feeling is a negative image of the
water's motion. And you may have noticed this sort of
(08:11):
illusion if you're a runner. So here's what normally happens.
Your body sends motor commands to the legs. It says, okay,
move fast, and correlated with your running is a visual
sensation where the visual world is flowing by your eyes
as a result. But when you run on a treadmill
(08:32):
at the gym, your brain is sending the same signals
to your legs. Run fast, but now your visual system
doesn't get the world flowing past. Instead, you're looking at
the gym wall in front of you the whole time.
So now when you step off, you experience the treadmill illusion.
(08:52):
With each step you take toward the locker room, the
world seems to stream by at a faster pace. It
looks as though you're moving forward more quickly than you
really are. Why does it happen Because your brain has
an expectation about how the act of moving your legs
should translate to the flow of the visual scene past
(09:14):
your eyes, and now after the treadmill, it's had to
adjust that relationship. So now there's an after effect. This
is just like with Aristotle's horse or the waterfall or
the feeling that you're still rocking after you get off
the boat. Your brain is readjusting its expectations about the world.
(09:34):
In the case of the treadmill, it's how the active
moving your legs should translate to the flow of the
visual scene past your eyes, and when you go back
to the normal world, there's an after effect. As another example,
you've surely noticed color after effects. If you stare at
a red dot for a little while and then you
look somewhere else and blink, you'll see the dot, but
(09:57):
now it looks green. Or if you stare at something
bright yellow for a little while and then you look away,
you'll see the opposite color blue. And these types of
color after effects can be surprisingly sophisticated. So consider something
called the McCullough after effect, named after the researcher Celeste McCulloch,
who discovered this in the nineteen sixties. So really get this.
(10:20):
I want you to imagine a bunch of black and
white lines like black white, black white, black white. Now
imagine a group of these that are horizontal and a
separate group of these that are vertical. Now, this can't
just be imagined as an illusion, but has to be
looked at to work. So please find a picture of
these lines on eagleman dot com Slash podcast to go
over there and take a look at this. Okay, So
(10:42):
look farther down the page and you'll see lines that
look the same as these black and white lines, but
now they're colored. The horizontal lines are green and the
vertical lines are red. And what I want you to
do is stare at these colored lines for a bit,
like a few minutes. The green lines go side side,
the red lines go up and down. Now go back
(11:04):
to the original black and white lines and you'll see
something amazing. You'll see that the spaces between the black
and white horizontal lines now look reddish, and the spaces
between the black and white vertical lines look greenish. This
is the McCullock effect. You gotta try this for yourself.
It's really worthwhile. Now why does it happen. It's because
(11:28):
when you stared at the colored lines, your brain said, hey, wait,
for some strange reason, everything horizontal is appearing green to
me and everything vertical is looking red to me. But
those features shouldn't really be correlated, and so it adjusts
itself to cancel out this strange relationship. Going on, and
(11:49):
when you look at the black and white lines again,
you're seeing the after effect. The horizontal lines were being
internally shifted in your brain towards the opposite color red,
and the vertical lines toward green. And again, I just
want to be clear, this has nothing to do with fatigue,
because in nineteen seventy five, two researchers showed that if
(12:11):
you stare at these red and green lines for fifteen minutes,
the after effect can last three and a half months.
So your brain is always doing an active recalibration of
the world out there. And this active recalibration is why
in the nineteen eighties many people began to see the
(12:31):
text in the books appearing to have a red tint
because at that time the population had just begun to
use computer monitors to do word processing. And as some
of you remember, unlike modern monitors, these early monitors only
displayed one color green, and so you typed a bunch
(12:51):
of lines of text on the screen and it looked
like horizontal lines of green on a black background. So
people would say there at these horizontal green rows for
hours at a time, all day while they were doing
their computer work. And so when they picked up a book,
the horizontal lines of black and white text were shaded
(13:14):
with the complementary color red. It was the McCullock effect.
Their brains were adjusting to a world of horizontal green lines,
and their reality changed accordingly. Now I mentioned that the
computer users in the nineteen eighties also experienced this illusion
when they looked at the IBM logo emblazoned on the
(13:36):
front of the floppy disk sleeve. If any of you
remember the old IBM logo, it's the three letters spelled
out of horizontal lines, as though you're looking at IBM
put through a egg slicer, And so the logo, just
like the horizontal lines in the book, looked tinged with red,
and designers at IBM were flummixed about this because they
(13:58):
had definitely not printed they're black and white design with
any red in it, and yet everyone was insisting that
they had. So all these after effects teach us something wild.
Even though you just open your eyes and look around
the world and assume you're just seeing reality out there,
in fact, your brain is doing massive work behind the scenes.
(14:20):
It's always recalibrating itself on the fly to try to
improve what it's seeing, to try to get rid of
information that doesn't matter, so it can surface the information
that does matter. And this active recalibration of the world
that's happening behind the scenes. This applies to everything. How
much motion is in the world, how stable the ground is,
(14:43):
whether vision flows past us when we move our legs,
whether lions are infused with color. None of the stuff
is decided in our genetics. It's all calibrated by our experience.
(15:12):
Now I want to take this concept of adaptation one
level deeper. I want to introduce the concept that your
brain is making things invisible if they are expected. So
take this as an example. Let's say you take a
yellow ping pong ball and you cut it exactly in half,
and you lay one hemisphere over each eye, and what
(15:33):
you'll see is the whole world is an even blanket
of yellow color. But within a few moments you don't
see any color at all. It's as though you're blind.
Your visual system just assumes that the world has become yellower,
and so it adapts so that you'll be sensitive to
other changes. So you should try this. You can find
(15:54):
any sort of totally featureless scene that's entirely one color
and fill your eyes with it. Like walk up to
a wall that's entirely painted red. Walk up real close.
Your eyes are filled with red, and you'll see the
color will quickly drain away to neutral. Your brain says, hey,
there's no more information here. I expect this color, So
(16:16):
I'm going to adjust things so that becomes invisible and
I'll tell you something else. The scene doesn't even have
to be featureless for things to fade away. So in
eighteen oh four, a Swiss physician named Ignos Troxler noticed
something really stunning, which is, if you stare at a
(16:37):
central point in the middle of a bunch of blobs
of color, all the busyness in the periphery, all those
blobs will eventually disappear. I have an example of this
on eagleman dot com slash podcast. Check it out. It's
really amazing. What you do is you keep your gaze
fixed right in the middle of the picture, and you'll
(16:57):
see there's all these smooth blobs of color, and within
about ten seconds, if you're keeping your eyes still, all
these surrounding blobs just start to disappear into the background,
and soon you are looking at what appears to be
a blank page. Now this is wild, and if you
move your eyes around just a bit, all these blobs
(17:19):
come back into your consciousness. This illusion is called the
Troxlur effect, and what it demonstrates is that if there's
a blob of something in your peripheral vision that is
not changing at all, it will evaporate from your perception. Now,
what is going on here? The answer is that your
visual system is always seeking motion and change. Something that's
(17:44):
fixed and none changing doesn't really matter, and it quickly
becomes invisible. Good information is expected to update, and things
that don't change get ignored by the system. So what
prevents your kitchen, your workplace from becoming like a Troxler
illusion with all the motionless features disappearing. Well, First, most
(18:07):
of the world is made up of hard edges, not blobs,
and those are easier for your visual system to hang onto.
But there's a deeper reason, although you're not generally aware
of it. Your eyes are constantly jumping and jiggling around.
So just observe a friend's eyeballs. You'll notice that her
(18:28):
eyeballs are making about three rapid jumps every second of
her waking life. These are called thecods. Bang bang bang bang, bang,
And if you watch even more closely, you'll find that
in between the big jumps, her eyes are constantly performing
these little micro jitters. They're called microsucods. Bang. Now, is
(18:50):
something wrong with your friend's eyes? No, these rapid movements,
both the big and the small. What they're doing is
they're keeping the image on the retina fresh totally unconsciously,
her eyes are working to maintain a constantly changing image.
Why do they bother. It's because any image that remains
(19:12):
perfectly fixed on one position on the retina will become invisible.
Here's how to prove this to yourself. If you wear
contact lenses, take a marker and draw a small shape
on the front of your contact right in the middle.
When you put the contact back in place on your eye,
you'll see something there, but it won't last long. It
(19:32):
rapidly fades to invisibility. Now, this phenomenon underscores the fundamental
fact that brains care about change. Just like with the
Troxler effect, features that don't change yield little information about
the world. All the important information comes from things in flux. Now,
(19:55):
if you don't have contact lenses, don't worry, because you're
already performing a similar experiment without knowing it. So you
have blood vessels that sit on top of your retina
at the back of your eye, and these blood vessels
should be seen superimposed on top of everything you look
at because they're smack in front of your photoreceptors. But
(20:18):
these blood vessels are totally invisible to your perception. Just
like the drawing on your contact lens. These blood vessels
are fixed in position with respect to your retina, So
no matter how much your eyes move, they can't refresh
the image of these blood vessels. Even though the vessels
(20:39):
interpose themselves between you and the world, they perceptually disappear.
It's like a magic trick. Now you might have noticed
a flash of these vessels when the eye doctor shines
the pen light in your eyes. Boom, there they are.
In this situation, the beam of light can cause the
(20:59):
vessel to cast a shadow at an unusual angle, and
your visual system suddenly takes notice something unexpected has just
occurred at the retina, and that's the only time you
witness this massive network that obstructs your view. If you
haven't seen this before, pause this podcast. Go into a
dark room and shine a beam of light in your eye.
(21:21):
From an angle, you'll see the blood vessels appear in
front of you. Just note that your visual system will
adapt fairly quickly, so the trick is to keep moving
the light to different angles to maintain the image. So
this strategy of ignoring the unchanging keeps your brain poised
to detect anything that moves or shifts or transforms at
(21:46):
the extreme. This is how reptile visual systems work. A
reptile can't see you if you stand still because they
only register change. They don't bother with position, and such
a system is perfectly sufficient. Reptiles have been surviving and
thriving for tens of millions of years. You probably remember
(22:06):
the scene in Jurassic Park where they say don't move.
The dinosaurs can't see you if you don't move. Well,
this is why reptiles ignore anything that does not change.
So let's return to the waterfall illusion. Why doesn't your
visual system shift so much that the waterfall is perceived
(22:26):
as standing still. Well, first, there might be limits to
the recalibration. It simply can't recalibrate enough to subtract off
the massive motion of the falls. But I think there's
another possibility, which is that you haven't watched the waterfall
for long enough, and if you did, it would eventually
recalibrate all the way. How long might that take? Two
(22:48):
months of steering at the waterfall two years. In theory,
if you were to watch for a long enough time,
then the short term changes in your visual system would
eventually to longer lasting changes, leading eventually to changes at
the deepest levels of the system. So ever present background
(23:09):
motion would become invisible to us. And this leads to
a crazy speculation that I made in my book Live Wire.
It's crazy, but logically sound. Are parts of the world
invisible to us that should be obvious. Imagine there was
something like a cosmic rainfall that had existed your entire life.
(23:31):
It would be completely invisible to you because you've never
seen otherwise. So your visual system would have set the
downward motion as it's zero point. If the cosmic rain
suddenly stopped, it would seem as though the whole world
were suddenly moving upward. We would believe that something had
just appeared ascending rain, even though the real rain had
(23:55):
just ended. And this situation could happen in any sensory channel.
Imagine the beat beep of a cosmic alarm clock with
no snooze button all the time, all throughout the cosmos
bee beep beep. If it were totally regular, you wouldn't
hear it because your brain would have adapted to it.
If the cosmic alarm suddenly ceased, everyone would hear a
(24:20):
great beat, beat, beat, but we would have no idea
that we were experiencing the after effect with the external
sound totally inside our heads. Successful adaptation makes regularities invisible.
(24:53):
So we've been talking about these illusions as the results
of adaptation, but there's actually the other way to look
at this. These illusions result from prediction. So if you
subtract away the downward motion of the waterfall or the
rocking of the boat, or the drawing on your contact lens,
(25:13):
that is equivalent to predicting its continued existence. In other words,
when brain circuitry adjusts, it's making a guess about what
the world is likely to be in the next moment.
It stops talking about news that is expected to continue.
So think again about your retinal blood vessels. They are
(25:35):
perceptually invisible because your visual system predicts them away. It
knows they are going to be there, so it ignores them.
It's only if those expectations are violated, like when a
light shines in from a strange angle that your brain
spends any energy on representing that data. Your brain doesn't
(25:58):
want to pay the energy car of spiking neurons, so
the goal is to reconfigure the network to waste as
little power as possible. If a pattern streams in that's
predictable or even partially guessable, the system saves energy by
structuring itself around that input, so it is not to
(26:20):
be surprised by it. If your brain is quieter, that
means there are fewer violations of expectations. Things in the
outside world are going approximately as forecast. In other words,
an energy conscious brain wants to predict away everything possible
so it can save its energy for just representing the unexpected.
(26:45):
Silence is golden. So although many neuroscientists think of activity
in neurons as the representation of things in the world,
it may turn out that spikes are the unpredicted, energy
expensive part. The representation of something totally expected would be
(27:06):
nothing but a hush falling over the neuronal forest. The
system makes adjustments only when it gets surprised. If your
brain thinks that all bricks weigh the same amount, and
then you attempt to pick up a brick made of lead.
The violation of your expectation causes cascades of changes to
(27:27):
deal with this new turn of events. But in contrast,
if everything is successfully predicted, there's no need to change anything.
For these reasons. When you first look at the Troxler picture,
you notice the blobs, and when you first put in
the contact lens you detect the drawn on shape. But
after a short time, your brain adjusts itself. It's no
(27:51):
longer surprised. Let me give you another example of the
brain predicting things away. So I talked in a previous
episode about the neosensory wristband, which converts sounds into patterns
of vibration on the skin, and deaf people wear the wristband,
and with time they can start to hear the world
(28:12):
through the signals that get to their brain via the skin.
If you're interested in learning more, that was episode twelve.
But the point I want to make here is what
happens when a hearing person first puts on the wristband.
They always react with surprise when they're feeling all the
sounds of the world, and they say WHOA, this thing
is picking up on my own voice. They're always startled
(28:33):
by that because it seems like they shouldn't be registering
their own speech. But of course, your ears pick up
on your voice all the time. It's typically the loudest
voice in your conversations because your own mouth is the
closest one to your ears. But because you can perfectly
predict your own vocalizations, you hardly hear your own voice.
(28:58):
You don't notice it your brain is predicting it away.
Or take this example, when people wear the wristband, they're
struck by the volume of other predictable sounds that they
normally pay no attention to because they're the ones who
create the sounds, like flushing the toilet or closing the
door behind them, or their own footsteps. It's not that
(29:20):
your auditory system doesn't register these sounds, but instead it's
that you actively predict them away. So you flush the
toilet or close the door or walk and these things
don't sound particularly loud to you, but it becomes obvious
when you're wearing the wristband. You can't believe how loud
these events are because your brain has not yet learned
(29:42):
to predict the signals coming up your arm, so your
brain actively recalibrates because that allows it to burn less energy.
But there's an even deeper principle at work here in
the darkness of your skull. You your brain is striving
to build an internal model of the outside world. When
(30:06):
you walk around your house, you don't pay any attention
to the environment because you already have a good model
of it. In contrast, when you're driving in a foreign
city you're trying to find a way to a particular hotel,
you're forced to look around at everything, the street signs,
the store naims, the building numbers because you don't already
have a good model of what to expect. So how
(30:30):
do you build up a good internal model? What is
the neural technology that allows you to zoom in on
those data points that don't match your expectation while ignoring
everything else that's already counted for. We call this attention.
You pay attention to the unexpected bang, or the unforeseen
(30:52):
brush on your skin, or the surprising movement in your periphery.
Attending allows you to put your high resolution sensors on
the problem and figure out how to incorporate it into
your model. Ah that's just the lawnmower. Ah, that's the kitten. Ah,
that's the housefly. Your model is now updated. In contrast,
(31:14):
you don't pay any attention to the feeling of the
shoe on your left foot, because you already have an
internal model of it, and that model is consistently predicting
what you're receiving, at least until you get a pebble
in your shoe that draws your attention because suddenly the
model calls for an update. The difference between predictions and
(31:38):
outcomes is the key to understanding a strange property of learning,
which is that if you're predicting something perfectly, your brain
doesn't need a change. Further so, say you learn that
the ding of your phone predicts that you just got
a text message. Your brain will quickly learn the relationship
between those two, in large part because of the relevance
(32:00):
of text messages to your social life. Then let's say
your phone gets a software update, and as a consequence,
the arrival of the text message is now signaled by
a ding plus a vibration. It turns out that your
brain won't train up on the vibration. This is an
effect known as blocking. Your brain already knows that the
(32:23):
ding predicts the text, so it has no need to
learn about something new. If your phone merely vibrates without
the ding, your brain won't know the meaning of that queue.
It has learned nothing about that. This phenomenon of blocking
makes sense when we understand that changes in the brain
(32:43):
happen only when there's a difference between what was expected
and what actually happened. So having an internal model of
the world allows us to make predictions and quickly detect
when we're wrong, which tells us where to attend and
how to update things. And this sort of system, by
the way, is becoming interesting to engineers thinking about the
(33:07):
future of machinery. Several companies are starting to work on
devices that operate this way, from tractors to airplanes. An
internal model of the world allows a machine to make
its best guesses about the events that are expected to unfold,
and when events are consistent with the predictions of the
(33:28):
machine's algorithm, nothing has to change, and it's only when
the inputs go off script that the software needs to
pay attention to update the machines model. Okay, so now
that we've talked about all the ways that your system adapts,
it will now be easy to understand how drugs modify
(33:48):
nervous systems. When you consume a drug, that changes the
number of receptors for that drug in the brain to
such an extent that you can look at a brain
after a person has died and determine his addictions by
gauging his molecular changes. This is why people become desensitized
(34:10):
or tolerant to a drug. The brain comes to predict
the presence of the drug. It adapts its receptor expression
so it can maintain a stable equilibrium when it receives
the next hit. In a physical literal way, the brain
comes to expect the drug to be there. The biological
(34:33):
details have calibrated themselves accordingly, and because the system now
predicts a certain amount to be present, more of the
drug is needed to achieve the original high. This recalibration
that's the basis of the ugly symptoms of drug withdrawal.
The more the brain is adapted to the drug, the
(34:56):
harder the fall is when the drug is taken away
with draw All symptoms vary by a drug, from sweating
to shakes to depression, but what they all have in
common is a powerful absence of something that was anticipated.
I suggested in my book Live Wire that this understanding
(35:16):
of neural predictions also gives us an understanding of heartbreak.
And this is because the people that you love become
part of you, not just metaphorically but physically. You absorb
people into your internal model of the world. Your brain
refashions itself around the expectation of their presence. So after
(35:42):
the breakup with a lover, or the death of a friend,
or the loss of a parent, the sudden absence represents
a major departure from homeostasis. As Khalil Jabraun put it
in the prophet quote and ever has a bin that
love knows not its own depth until the hour of separation.
(36:06):
In this way, your brain becomes like a negative image
of everyone you've ever come in contact with. Your lovers,
your friends, your parents. They fill in their expected shapes,
and just like feeling the waves after you've departed the
boat or craving the drug when it's not there, so
(36:27):
your brain calls for the people in your life to
be there. And when someone moves away or rejects you,
or dies, your brain struggles with its thwarted expectations. Slowly
through time, it has to readjust to a world without
(36:47):
that person. Now I want to cover one more thing
in today's episode, which is how your brain knows how
to adjust. And I suggest it in Live Wire that
there's a very simple but surprise strategy underlying all this.
And to understand this, consider something like phototropism in plants.
(37:07):
This is where plants capture maximum light by constantly adjusting
their position. So if you watch a plant growing in
fast motion, you'll see that it doesn't grow straight towards
the light source. Instead, it overshoots its trajectory by a
little bit, then it undershoots by a bit, and back
and forth. It's not a pre planned mission. It's a
(37:30):
spastic dance with constant correction. And by the way, you
find something similar in the strategy that bacteria use to
find food. When they're searching for the center of a
food source, like a bit of sugar that's fallen on
the kitchen counter, they make their way to the sugar
by employing three elegantly simple rules. One randomly selected direction
(37:56):
and move in a straight line. Two if things are
getting better, keep going. Three if things are getting worse,
randomly change directions by tumbling. In other words, the strategy
is to lock down the approach when conditions are improving
and dump it when it's not working. By this simple policy,
(38:18):
a bacterium can quickly and efficiently work its way to
the densest point of the food source. So I've proposed
that there's a similar principle at work in the brain.
Instead of working its way towards maximizing sunlight or food,
it works towards maximizing information. And I call this strategy infotropism.
(38:42):
The hypothesis suggests that neural circuitry constantly shifts to maximize
the amount of information it can extract from the world.
So consider what we've heard in some of the previous episodes.
We saw the way the brain comes to figure out
its sensory organs, its eyes and ears and nose and fingertips.
(39:02):
When you're a baby, you don't know what those data
streams mean, and your brain figures it out with experience.
And we saw the way a brain comes to learn
how to control a body, no matter if that body
possesses fins or legs or robotic arms, whatever the case.
The brain fine tunes its circuitry to maximize data it
(39:23):
streams from the world. So this fine tuning is helped
along by rewards, which cause broadcasts throughout the circuitry to
announce that something worked. In this way, with a minimum
of preprogramming, the system works out how to optimize its
interaction with the world. Or Another example is that your
(39:44):
brain will adjust to whatever language you're exposed to as
a baby, such that you can hear those sounds, but
you lose the ability to distinguish other sounds and other languages.
This is all infotropism at work. Baby's brains adjust to
maximize the data that matters around them. On a longer
(40:07):
time scale, when a person goes blind, other senses take
over the visual cortex, and in an upcoming episode we'll
learn how neurons actually pull this off, but for now,
note that we can interpret this takeover as in photropism.
The brain is maximizing its resources so that it can
best interpret whatever data flows in. And just think about
(40:30):
the Mecullic effect with the horizontal and vertical colored lines.
Your visual system works to separate the dimensions of color
and orientation because it's trying to maximize information from the world,
and so accordingly, it doesn't want to mix together these
separable measures. So although the effect is typically viewed as
(40:52):
simply a fun visual illusion. The work happens under the
hood for a deeper reason. If something were causing a
tinge to appear on lines, like some strange overhead lighting
or something wrong with your optics, your brain would reorganize
itself to take care of this, canceling out this weird relationship,
(41:13):
and by doing so, it would maximize your capacity to
extract information about colors and orientations separately. By separating two
dimensions that statistically should be unyoked, your brain can best
gather information from the world. Here's a cool example of
infotropism at the level of neurons, the retina in your eye,
(41:37):
the set of photoreceptors the back of your eye captures
the light of the world and tells the brain what
it's seeing out there. But it reads the world differently
in the day and in the night. So in bright
daylight there are plenty of photons to capture, and so
each photoreceptor just takes care of its own tiny dot
(41:58):
of the scene, and this gives nice and high resolution.
But at night it's a different story. There are few
photons to be captured, and so now the important job
of the retina is to detect that something was there,
even if it can't tell that with high resolution. So
at night, the photoreceptors do their work very differently. Essentially,
(42:21):
they join forces with their neighbors. They change the details
of their internal molecular cascades so they can link arms
with one another. Now they're operating as a team, they
can detect much lower levels of light, they can detect
fewer photons out there. So what we have here is
this beautiful and sophisticated strategy that has the retina operate
(42:46):
differently as the light levels go up or down. When
it's bright out, the system achieves high spatial resolution, and
when it's dark, photoreceptors pool together to have a better
chance to catch photons, which results in vision that's more
sensitive to dim light but blurrier in resolution. You've surely
(43:08):
noticed this at nighttime when you can't quite make out
what something is out there, whereas it would be easy
in the daytime. So the point here is that your eyes,
your retinas, put enormous work into shifting themselves around to
maximize information that matters whether the photons are plentiful or
they're rare. The retina optimizes itself to get useful data.
(43:33):
In the day, it captures the most detail so that
it can spot the rabbit at the distance. In dim light,
it shifts to higher sensitivity to capture whatever's out there
with lower detail, capturing the shadowy essence of the jaguar
lurking in the gloom. Mother Nature figured out not only
how to build an eye, but also how to adjust
(43:56):
its circuitry on the fly, so it can operate well
in different contexts, all to make the best use of
what's available. It is infotropic. Just like the plant seeks light,
the retina seeks data, and more generally, as we've seen
in this episode, the brain is always seeking information. It
(44:18):
constantly adjusts its circuitry to maximize the data it can
draw from the world. So if you're staring at a waterfall,
it discounts the downward motion so it can be sensitive
to other things. If you put on yellow sunglasses, your
brain discounts that shift in color so that it can
be more sensitive to other information. And you can detect
(44:41):
this shift when you take off the sunglasses and you
have a big blue color after effect. So to wrap up,
this episode plugs into the big picture we've been talking
about in many episodes, and we'll see in many more,
which is that the brain is locked in silence and darkness,
and its job is to build an internal model of
the outside world. And what we're seeing a little more
(45:04):
clearly now is that its internal model is really there
to make predictions. If the world proceeds as expected, the
brain saves energy. So when the world changes, let's say,
becoming a place with lots of downward motion or yellow
light or something, the brain adjusts to match it so
that it can be maximumly positioned to gather new information
(45:28):
that's unexpected. So these after effect illusions seem funny and
may be trivial, but they're actually a powerful way for
us to reveal this constant adaptation that your brain is doing.
Over the course of many episodes coming up, we're going
to see how fundamentally the brain is a prediction machine,
(45:48):
and that is the driving engine behind its constant self reconfiguration.
By modeling the state of the world, the brain reshapes
itself to have good expectation and therefore to be maximally
sensitive to the unexpected. Go to Eagleman dot com slash
(46:10):
podcast for more information and to find further reading. Send
me an email at podcasts at eagleman dot com with
questions or discussions, and I'll be making an episode soon
in which I address those. Until next time, I'm David Eagleman,
and this is Inner Cosmos.
Why do you still feel the waves after you get
off a boat and when you get off the treadmill
at the gym. Why does everything seem to be flowing
past you faster? What does heartbreak have in common with
drug withdrawal? And why after you stare at a waterfall
for a while, do the rocks on the side seem
to be crawling upward? Why did people in the nineteen
(00:27):
eighties think their book pages had some color red in them,
but no one thought that before or after the eighties?
And what does any of this have to do with
the great philosopher Aristotle watching a horse stuck in the river?
Welcome to Inner Cosmos with me David Eagleman. I'm a
(00:49):
neuroscientist and author at Stanford and in these episodes we
sail deeply into our three pound universe to understand why
and how our lives look the way they do. Today's
(01:09):
episode is about the way the brain readjusts its circuitry
on really fast timescales, and why your brain is constantly
doing this. In the nineteen eighties, tens of thousands of
people began to notice something really weird when they looked
at a floppy disk envelope with the black and white
(01:31):
IBM logo on the front, the letters IBM seemed to
have a red tint, And the same thing happened when
people looked at pages in a book, the text seemed
like it was shaded red. Now, if that's not weird enough,
check this out. This only happened in the nineteen eighties.
People didn't perceive a red tint before or after this decade.
(01:54):
So what was changing about brains during that window. I'll
tell you the answer. But to understand this, we're first
going to step back even further in history, twenty four
hundred years further to Aristotle. Now, Aristotle, as you know,
is a philosopher and a polymath whose intellectual interests were boundless.
(02:18):
He wrote about essentially every subject you could imagine. But
we're coming to Aristotle not because of something he wrote,
but instead because he was insightful enough to pause and
notice when something didn't make sense. What happened is that
someone was trying to cross a fast flowing river on
his horse, and the horse got stuck halfway in the river.
(02:41):
So Aristotle, along with some other people, ended up watching
this rescue operation. The way you might have a bunch
of people looking at a car crash nowadays. So you
have all these Greeks all standing there in their togas,
talking and chatting and watching the horse in the fast
flowing river. Now most of the people were presumably just
talking about stuff that made no difference to history. But
(03:03):
Aristotle stood at the side of the river intently watching
this horse. Now Here was the key about Aristotle, a
general characteristic that made him so famous to history. He
was a good observer. He paid attention. He noticed when
things were unusual, and when he found something, he would
(03:25):
sink his teeth into the problem and wouldn't let go.
And what he noticed here was something that possibly the
others did too, but they ignored it, and they didn't
think any further on it. What he noticed is that
once the horse got unstuck and the man made it
to the shore, and Aristotle looked away, he noticed that
everything on the land, the rocks, the huts, the trees,
(03:47):
everything seemed to be drifting, and they all seemed to
be moving in the opposite direction of the river. Now
you may have seen something like this too. The easiest
way to experience Aristotle's confusion and delight is to stare
at a waterfall and after you keep your eyes locked
on it for a while, look over to the rocks
(04:10):
on the side of the waterfall, and the rocks appear
to be moving upward. Now, this illusion that he first
made note of twenty four hundred years ago has come
to be known as the motion after effect. Why does
it happen. The activity of particular neurons in your visual
cortex represents downward motion, and the activity of other neurons
(04:35):
represents upward motion. And they're always locked in battle, and
most of the time the competition is evenly pitched, and
they evenly inhibit each other, and as a result, the
world appears to you to be moving neither up nor down. So,
given this, a popular explanation for the motion after effect
(04:56):
is fatigue. By staring at the downward motion of the waterfall,
you burn up a good deal of energy in your
downward coating neurons, and now their vigor is temporarily depleted.
So the battle tips in favor of the upward encoding neurons,
and as a result of this unbalanced activity, you perceive
(05:19):
that the rocks are climbing upward. You have this net
movement upward. Now, this fatigue hypothesis is very appealing in
its simplicity, but it's wrong. Why because it can't explain
some critical facts about the illusion. So imagine you watch
the downward waterfall for a while and then you close
(05:41):
your eyes tightly, say, for three hours you close your eyes.
When you reopen your eyes, you'll see the rocks crawling upward.
And what that tells us is it's not about a
temporary energy depletion in neurons. There's something deeper going on here.
And here's what that deeper thing is is the illusion
(06:01):
comes about not because of passive fatigue, but instead because
of an active recalibration in your brain. Your visual system
is exposed to continuous downward motion, and after a while
it comes to assume this is the new normal. At first,
downward motion is dramatic information to your brain, but after
(06:25):
a while of staring, you receive no new information from
downwardness here. So, as far as your brain is concerned,
this is the new reality, a world that flows more
down than up. So your visual system carefully rebalances its
expectations to mirror the world to expect more downward than
(06:47):
upward activity. Now, when you look away from the waterfall
and toward the cliff side. This recalibrated set point becomes
obvious because now rocks and trees are flowing upward toward
the sky. Your set point has shifted. In other words,
what counts as standing still has shifted. Why well, your
(07:10):
brain always wants to set up a ground truth so
that it can be better at detecting change. In this case,
when your visual field is filled with the site of
the waterfall, your brain is working to subtract off all
that downward motion. All that downward motion is no longer
(07:30):
informative in the way that it was, and so the
circuitry it adjusts itself so that it can be maximumly
sensitive to new information. Now, if you watch for it
the way Aristotle would, you'll see that this kind of
recalibration happens to you all the time. Think about when
you're on a small boat on the ocean. You are
(07:51):
rocking with the waves for a while, and then when
you get off the boat, the land seems to be
rocking for a while. It feels as though you're still
on the water. Now you may have thought about it
this way, Hey, my body thinks it's still on the water.
What you're actually feeling is a negative image of the
water's motion. And you may have noticed this sort of
(08:11):
illusion if you're a runner. So here's what normally happens.
Your body sends motor commands to the legs. It says, okay,
move fast, and correlated with your running is a visual
sensation where the visual world is flowing by your eyes
as a result. But when you run on a treadmill
(08:32):
at the gym, your brain is sending the same signals
to your legs. Run fast, but now your visual system
doesn't get the world flowing past. Instead, you're looking at
the gym wall in front of you the whole time.
So now when you step off, you experience the treadmill illusion.
(08:52):
With each step you take toward the locker room, the
world seems to stream by at a faster pace. It
looks as though you're moving forward more quickly than you
really are. Why does it happen Because your brain has
an expectation about how the act of moving your legs
should translate to the flow of the visual scene past
(09:14):
your eyes, and now after the treadmill, it's had to
adjust that relationship. So now there's an after effect. This
is just like with Aristotle's horse or the waterfall or
the feeling that you're still rocking after you get off
the boat. Your brain is readjusting its expectations about the world.
(09:34):
In the case of the treadmill, it's how the active
moving your legs should translate to the flow of the
visual scene past your eyes, and when you go back
to the normal world, there's an after effect. As another example,
you've surely noticed color after effects. If you stare at
a red dot for a little while and then you
look somewhere else and blink, you'll see the dot, but
(09:57):
now it looks green. Or if you stare at something
bright yellow for a little while and then you look away,
you'll see the opposite color blue. And these types of
color after effects can be surprisingly sophisticated. So consider something
called the McCullough after effect, named after the researcher Celeste McCulloch,
who discovered this in the nineteen sixties. So really get this.
(10:20):
I want you to imagine a bunch of black and
white lines like black white, black white, black white. Now
imagine a group of these that are horizontal and a
separate group of these that are vertical. Now, this can't
just be imagined as an illusion, but has to be
looked at to work. So please find a picture of
these lines on eagleman dot com Slash podcast to go
over there and take a look at this. Okay, So
(10:42):
look farther down the page and you'll see lines that
look the same as these black and white lines, but
now they're colored. The horizontal lines are green and the
vertical lines are red. And what I want you to
do is stare at these colored lines for a bit,
like a few minutes. The green lines go side side,
the red lines go up and down. Now go back
(11:04):
to the original black and white lines and you'll see
something amazing. You'll see that the spaces between the black
and white horizontal lines now look reddish, and the spaces
between the black and white vertical lines look greenish. This
is the McCullock effect. You gotta try this for yourself.
It's really worthwhile. Now why does it happen. It's because
(11:28):
when you stared at the colored lines, your brain said, hey, wait,
for some strange reason, everything horizontal is appearing green to
me and everything vertical is looking red to me. But
those features shouldn't really be correlated, and so it adjusts
itself to cancel out this strange relationship. Going on, and
(11:49):
when you look at the black and white lines again,
you're seeing the after effect. The horizontal lines were being
internally shifted in your brain towards the opposite color red,
and the vertical lines toward green. And again, I just
want to be clear, this has nothing to do with fatigue,
because in nineteen seventy five, two researchers showed that if
(12:11):
you stare at these red and green lines for fifteen minutes,
the after effect can last three and a half months.
So your brain is always doing an active recalibration of
the world out there. And this active recalibration is why
in the nineteen eighties many people began to see the
(12:31):
text in the books appearing to have a red tint
because at that time the population had just begun to
use computer monitors to do word processing. And as some
of you remember, unlike modern monitors, these early monitors only
displayed one color green, and so you typed a bunch
(12:51):
of lines of text on the screen and it looked
like horizontal lines of green on a black background. So
people would say there at these horizontal green rows for
hours at a time, all day while they were doing
their computer work. And so when they picked up a book,
the horizontal lines of black and white text were shaded
(13:14):
with the complementary color red. It was the McCullock effect.
Their brains were adjusting to a world of horizontal green lines,
and their reality changed accordingly. Now I mentioned that the
computer users in the nineteen eighties also experienced this illusion
when they looked at the IBM logo emblazoned on the
(13:36):
front of the floppy disk sleeve. If any of you
remember the old IBM logo, it's the three letters spelled
out of horizontal lines, as though you're looking at IBM
put through a egg slicer, And so the logo, just
like the horizontal lines in the book, looked tinged with red,
and designers at IBM were flummixed about this because they
(13:58):
had definitely not printed they're black and white design with
any red in it, and yet everyone was insisting that
they had. So all these after effects teach us something wild.
Even though you just open your eyes and look around
the world and assume you're just seeing reality out there,
in fact, your brain is doing massive work behind the scenes.
(14:20):
It's always recalibrating itself on the fly to try to
improve what it's seeing, to try to get rid of
information that doesn't matter, so it can surface the information
that does matter. And this active recalibration of the world
that's happening behind the scenes. This applies to everything. How
much motion is in the world, how stable the ground is,
(14:43):
whether vision flows past us when we move our legs,
whether lions are infused with color. None of the stuff
is decided in our genetics. It's all calibrated by our experience.
(15:12):
Now I want to take this concept of adaptation one
level deeper. I want to introduce the concept that your
brain is making things invisible if they are expected. So
take this as an example. Let's say you take a
yellow ping pong ball and you cut it exactly in half,
and you lay one hemisphere over each eye, and what
(15:33):
you'll see is the whole world is an even blanket
of yellow color. But within a few moments you don't
see any color at all. It's as though you're blind.
Your visual system just assumes that the world has become yellower,
and so it adapts so that you'll be sensitive to
other changes. So you should try this. You can find
(15:54):
any sort of totally featureless scene that's entirely one color
and fill your eyes with it. Like walk up to
a wall that's entirely painted red. Walk up real close.
Your eyes are filled with red, and you'll see the
color will quickly drain away to neutral. Your brain says, hey,
there's no more information here. I expect this color, So
(16:16):
I'm going to adjust things so that becomes invisible and
I'll tell you something else. The scene doesn't even have
to be featureless for things to fade away. So in
eighteen oh four, a Swiss physician named Ignos Troxler noticed
something really stunning, which is, if you stare at a
(16:37):
central point in the middle of a bunch of blobs
of color, all the busyness in the periphery, all those
blobs will eventually disappear. I have an example of this
on eagleman dot com slash podcast. Check it out. It's
really amazing. What you do is you keep your gaze
fixed right in the middle of the picture, and you'll
(16:57):
see there's all these smooth blobs of color, and within
about ten seconds, if you're keeping your eyes still, all
these surrounding blobs just start to disappear into the background,
and soon you are looking at what appears to be
a blank page. Now this is wild, and if you
move your eyes around just a bit, all these blobs
(17:19):
come back into your consciousness. This illusion is called the
Troxlur effect, and what it demonstrates is that if there's
a blob of something in your peripheral vision that is
not changing at all, it will evaporate from your perception. Now,
what is going on here? The answer is that your
visual system is always seeking motion and change. Something that's
(17:44):
fixed and none changing doesn't really matter, and it quickly
becomes invisible. Good information is expected to update, and things
that don't change get ignored by the system. So what
prevents your kitchen, your workplace from becoming like a Troxler
illusion with all the motionless features disappearing. Well, First, most
(18:07):
of the world is made up of hard edges, not blobs,
and those are easier for your visual system to hang onto.
But there's a deeper reason, although you're not generally aware
of it. Your eyes are constantly jumping and jiggling around.
So just observe a friend's eyeballs. You'll notice that her
(18:28):
eyeballs are making about three rapid jumps every second of
her waking life. These are called thecods. Bang bang bang bang, bang,
And if you watch even more closely, you'll find that
in between the big jumps, her eyes are constantly performing
these little micro jitters. They're called microsucods. Bang. Now, is
(18:50):
something wrong with your friend's eyes? No, these rapid movements,
both the big and the small. What they're doing is
they're keeping the image on the retina fresh totally unconsciously,
her eyes are working to maintain a constantly changing image.
Why do they bother. It's because any image that remains
(19:12):
perfectly fixed on one position on the retina will become invisible.
Here's how to prove this to yourself. If you wear
contact lenses, take a marker and draw a small shape
on the front of your contact right in the middle.
When you put the contact back in place on your eye,
you'll see something there, but it won't last long. It
(19:32):
rapidly fades to invisibility. Now, this phenomenon underscores the fundamental
fact that brains care about change. Just like with the
Troxler effect, features that don't change yield little information about
the world. All the important information comes from things in flux. Now,
(19:55):
if you don't have contact lenses, don't worry, because you're
already performing a similar experiment without knowing it. So you
have blood vessels that sit on top of your retina
at the back of your eye, and these blood vessels
should be seen superimposed on top of everything you look
at because they're smack in front of your photoreceptors. But
(20:18):
these blood vessels are totally invisible to your perception. Just
like the drawing on your contact lens. These blood vessels
are fixed in position with respect to your retina, So
no matter how much your eyes move, they can't refresh
the image of these blood vessels. Even though the vessels
(20:39):
interpose themselves between you and the world, they perceptually disappear.
It's like a magic trick. Now you might have noticed
a flash of these vessels when the eye doctor shines
the pen light in your eyes. Boom, there they are.
In this situation, the beam of light can cause the
(20:59):
vessel to cast a shadow at an unusual angle, and
your visual system suddenly takes notice something unexpected has just
occurred at the retina, and that's the only time you
witness this massive network that obstructs your view. If you
haven't seen this before, pause this podcast. Go into a
dark room and shine a beam of light in your eye.
(21:21):
From an angle, you'll see the blood vessels appear in
front of you. Just note that your visual system will
adapt fairly quickly, so the trick is to keep moving
the light to different angles to maintain the image. So
this strategy of ignoring the unchanging keeps your brain poised
to detect anything that moves or shifts or transforms at
(21:46):
the extreme. This is how reptile visual systems work. A
reptile can't see you if you stand still because they
only register change. They don't bother with position, and such
a system is perfectly sufficient. Reptiles have been surviving and
thriving for tens of millions of years. You probably remember
(22:06):
the scene in Jurassic Park where they say don't move.
The dinosaurs can't see you if you don't move. Well,
this is why reptiles ignore anything that does not change.
So let's return to the waterfall illusion. Why doesn't your
visual system shift so much that the waterfall is perceived
(22:26):
as standing still. Well, first, there might be limits to
the recalibration. It simply can't recalibrate enough to subtract off
the massive motion of the falls. But I think there's
another possibility, which is that you haven't watched the waterfall
for long enough, and if you did, it would eventually
recalibrate all the way. How long might that take? Two
(22:48):
months of steering at the waterfall two years. In theory,
if you were to watch for a long enough time,
then the short term changes in your visual system would
eventually to longer lasting changes, leading eventually to changes at
the deepest levels of the system. So ever present background
(23:09):
motion would become invisible to us. And this leads to
a crazy speculation that I made in my book Live Wire.
It's crazy, but logically sound. Are parts of the world
invisible to us that should be obvious. Imagine there was
something like a cosmic rainfall that had existed your entire life.
(23:31):
It would be completely invisible to you because you've never
seen otherwise. So your visual system would have set the
downward motion as it's zero point. If the cosmic rain
suddenly stopped, it would seem as though the whole world
were suddenly moving upward. We would believe that something had
just appeared ascending rain, even though the real rain had
(23:55):
just ended. And this situation could happen in any sensory channel.
Imagine the beat beep of a cosmic alarm clock with
no snooze button all the time, all throughout the cosmos
bee beep beep. If it were totally regular, you wouldn't
hear it because your brain would have adapted to it.
If the cosmic alarm suddenly ceased, everyone would hear a
(24:20):
great beat, beat, beat, but we would have no idea
that we were experiencing the after effect with the external
sound totally inside our heads. Successful adaptation makes regularities invisible.
(24:53):
So we've been talking about these illusions as the results
of adaptation, but there's actually the other way to look
at this. These illusions result from prediction. So if you
subtract away the downward motion of the waterfall or the
rocking of the boat, or the drawing on your contact lens,
(25:13):
that is equivalent to predicting its continued existence. In other words,
when brain circuitry adjusts, it's making a guess about what
the world is likely to be in the next moment.
It stops talking about news that is expected to continue.
So think again about your retinal blood vessels. They are
(25:35):
perceptually invisible because your visual system predicts them away. It
knows they are going to be there, so it ignores them.
It's only if those expectations are violated, like when a
light shines in from a strange angle that your brain
spends any energy on representing that data. Your brain doesn't
(25:58):
want to pay the energy car of spiking neurons, so
the goal is to reconfigure the network to waste as
little power as possible. If a pattern streams in that's
predictable or even partially guessable, the system saves energy by
structuring itself around that input, so it is not to
(26:20):
be surprised by it. If your brain is quieter, that
means there are fewer violations of expectations. Things in the
outside world are going approximately as forecast. In other words,
an energy conscious brain wants to predict away everything possible
so it can save its energy for just representing the unexpected.
(26:45):
Silence is golden. So although many neuroscientists think of activity
in neurons as the representation of things in the world,
it may turn out that spikes are the unpredicted, energy
expensive part. The representation of something totally expected would be
(27:06):
nothing but a hush falling over the neuronal forest. The
system makes adjustments only when it gets surprised. If your
brain thinks that all bricks weigh the same amount, and
then you attempt to pick up a brick made of lead.
The violation of your expectation causes cascades of changes to
(27:27):
deal with this new turn of events. But in contrast,
if everything is successfully predicted, there's no need to change anything.
For these reasons. When you first look at the Troxler picture,
you notice the blobs, and when you first put in
the contact lens you detect the drawn on shape. But
after a short time, your brain adjusts itself. It's no
(27:51):
longer surprised. Let me give you another example of the
brain predicting things away. So I talked in a previous
episode about the neosensory wristband, which converts sounds into patterns
of vibration on the skin, and deaf people wear the wristband,
and with time they can start to hear the world
(28:12):
through the signals that get to their brain via the skin.
If you're interested in learning more, that was episode twelve.
But the point I want to make here is what
happens when a hearing person first puts on the wristband.
They always react with surprise when they're feeling all the
sounds of the world, and they say WHOA, this thing
is picking up on my own voice. They're always startled
(28:33):
by that because it seems like they shouldn't be registering
their own speech. But of course, your ears pick up
on your voice all the time. It's typically the loudest
voice in your conversations because your own mouth is the
closest one to your ears. But because you can perfectly
predict your own vocalizations, you hardly hear your own voice.
(28:58):
You don't notice it your brain is predicting it away.
Or take this example, when people wear the wristband, they're
struck by the volume of other predictable sounds that they
normally pay no attention to because they're the ones who
create the sounds, like flushing the toilet or closing the
door behind them, or their own footsteps. It's not that
(29:20):
your auditory system doesn't register these sounds, but instead it's
that you actively predict them away. So you flush the
toilet or close the door or walk and these things
don't sound particularly loud to you, but it becomes obvious
when you're wearing the wristband. You can't believe how loud
these events are because your brain has not yet learned
(29:42):
to predict the signals coming up your arm, so your
brain actively recalibrates because that allows it to burn less energy.
But there's an even deeper principle at work here in
the darkness of your skull. You your brain is striving
to build an internal model of the outside world. When
(30:06):
you walk around your house, you don't pay any attention
to the environment because you already have a good model
of it. In contrast, when you're driving in a foreign
city you're trying to find a way to a particular hotel,
you're forced to look around at everything, the street signs,
the store naims, the building numbers because you don't already
have a good model of what to expect. So how
(30:30):
do you build up a good internal model? What is
the neural technology that allows you to zoom in on
those data points that don't match your expectation while ignoring
everything else that's already counted for. We call this attention.
You pay attention to the unexpected bang, or the unforeseen
(30:52):
brush on your skin, or the surprising movement in your periphery.
Attending allows you to put your high resolution sensors on
the problem and figure out how to incorporate it into
your model. Ah that's just the lawnmower. Ah, that's the kitten. Ah,
that's the housefly. Your model is now updated. In contrast,
(31:14):
you don't pay any attention to the feeling of the
shoe on your left foot, because you already have an
internal model of it, and that model is consistently predicting
what you're receiving, at least until you get a pebble
in your shoe that draws your attention because suddenly the
model calls for an update. The difference between predictions and
(31:38):
outcomes is the key to understanding a strange property of learning,
which is that if you're predicting something perfectly, your brain
doesn't need a change. Further so, say you learn that
the ding of your phone predicts that you just got
a text message. Your brain will quickly learn the relationship
between those two, in large part because of the relevance
(32:00):
of text messages to your social life. Then let's say
your phone gets a software update, and as a consequence,
the arrival of the text message is now signaled by
a ding plus a vibration. It turns out that your
brain won't train up on the vibration. This is an
effect known as blocking. Your brain already knows that the
(32:23):
ding predicts the text, so it has no need to
learn about something new. If your phone merely vibrates without
the ding, your brain won't know the meaning of that queue.
It has learned nothing about that. This phenomenon of blocking
makes sense when we understand that changes in the brain
(32:43):
happen only when there's a difference between what was expected
and what actually happened. So having an internal model of
the world allows us to make predictions and quickly detect
when we're wrong, which tells us where to attend and
how to update things. And this sort of system, by
the way, is becoming interesting to engineers thinking about the
(33:07):
future of machinery. Several companies are starting to work on
devices that operate this way, from tractors to airplanes. An
internal model of the world allows a machine to make
its best guesses about the events that are expected to unfold,
and when events are consistent with the predictions of the
(33:28):
machine's algorithm, nothing has to change, and it's only when
the inputs go off script that the software needs to
pay attention to update the machines model. Okay, so now
that we've talked about all the ways that your system adapts,
it will now be easy to understand how drugs modify
(33:48):
nervous systems. When you consume a drug, that changes the
number of receptors for that drug in the brain to
such an extent that you can look at a brain
after a person has died and determine his addictions by
gauging his molecular changes. This is why people become desensitized
(34:10):
or tolerant to a drug. The brain comes to predict
the presence of the drug. It adapts its receptor expression
so it can maintain a stable equilibrium when it receives
the next hit. In a physical literal way, the brain
comes to expect the drug to be there. The biological
(34:33):
details have calibrated themselves accordingly, and because the system now
predicts a certain amount to be present, more of the
drug is needed to achieve the original high. This recalibration
that's the basis of the ugly symptoms of drug withdrawal.
The more the brain is adapted to the drug, the
(34:56):
harder the fall is when the drug is taken away
with draw All symptoms vary by a drug, from sweating
to shakes to depression, but what they all have in
common is a powerful absence of something that was anticipated.
I suggested in my book Live Wire that this understanding
(35:16):
of neural predictions also gives us an understanding of heartbreak.
And this is because the people that you love become
part of you, not just metaphorically but physically. You absorb
people into your internal model of the world. Your brain
refashions itself around the expectation of their presence. So after
(35:42):
the breakup with a lover, or the death of a friend,
or the loss of a parent, the sudden absence represents
a major departure from homeostasis. As Khalil Jabraun put it
in the prophet quote and ever has a bin that
love knows not its own depth until the hour of separation.
(36:06):
In this way, your brain becomes like a negative image
of everyone you've ever come in contact with. Your lovers,
your friends, your parents. They fill in their expected shapes,
and just like feeling the waves after you've departed the
boat or craving the drug when it's not there, so
(36:27):
your brain calls for the people in your life to
be there. And when someone moves away or rejects you,
or dies, your brain struggles with its thwarted expectations. Slowly
through time, it has to readjust to a world without
(36:47):
that person. Now I want to cover one more thing
in today's episode, which is how your brain knows how
to adjust. And I suggest it in Live Wire that
there's a very simple but surprise strategy underlying all this.
And to understand this, consider something like phototropism in plants.
(37:07):
This is where plants capture maximum light by constantly adjusting
their position. So if you watch a plant growing in
fast motion, you'll see that it doesn't grow straight towards
the light source. Instead, it overshoots its trajectory by a
little bit, then it undershoots by a bit, and back
and forth. It's not a pre planned mission. It's a
(37:30):
spastic dance with constant correction. And by the way, you
find something similar in the strategy that bacteria use to
find food. When they're searching for the center of a
food source, like a bit of sugar that's fallen on
the kitchen counter, they make their way to the sugar
by employing three elegantly simple rules. One randomly selected direction
(37:56):
and move in a straight line. Two if things are
getting better, keep going. Three if things are getting worse,
randomly change directions by tumbling. In other words, the strategy
is to lock down the approach when conditions are improving
and dump it when it's not working. By this simple policy,
(38:18):
a bacterium can quickly and efficiently work its way to
the densest point of the food source. So I've proposed
that there's a similar principle at work in the brain.
Instead of working its way towards maximizing sunlight or food,
it works towards maximizing information. And I call this strategy infotropism.
(38:42):
The hypothesis suggests that neural circuitry constantly shifts to maximize
the amount of information it can extract from the world.
So consider what we've heard in some of the previous episodes.
We saw the way the brain comes to figure out
its sensory organs, its eyes and ears and nose and fingertips.
(39:02):
When you're a baby, you don't know what those data
streams mean, and your brain figures it out with experience.
And we saw the way a brain comes to learn
how to control a body, no matter if that body
possesses fins or legs or robotic arms, whatever the case.
The brain fine tunes its circuitry to maximize data it
(39:23):
streams from the world. So this fine tuning is helped
along by rewards, which cause broadcasts throughout the circuitry to
announce that something worked. In this way, with a minimum
of preprogramming, the system works out how to optimize its
interaction with the world. Or Another example is that your
(39:44):
brain will adjust to whatever language you're exposed to as
a baby, such that you can hear those sounds, but
you lose the ability to distinguish other sounds and other languages.
This is all infotropism at work. Baby's brains adjust to
maximize the data that matters around them. On a longer
(40:07):
time scale, when a person goes blind, other senses take
over the visual cortex, and in an upcoming episode we'll
learn how neurons actually pull this off, but for now,
note that we can interpret this takeover as in photropism.
The brain is maximizing its resources so that it can
best interpret whatever data flows in. And just think about
(40:30):
the Mecullic effect with the horizontal and vertical colored lines.
Your visual system works to separate the dimensions of color
and orientation because it's trying to maximize information from the world,
and so accordingly, it doesn't want to mix together these
separable measures. So although the effect is typically viewed as
(40:52):
simply a fun visual illusion. The work happens under the
hood for a deeper reason. If something were causing a
tinge to appear on lines, like some strange overhead lighting
or something wrong with your optics, your brain would reorganize
itself to take care of this, canceling out this weird relationship,
(41:13):
and by doing so, it would maximize your capacity to
extract information about colors and orientations separately. By separating two
dimensions that statistically should be unyoked, your brain can best
gather information from the world. Here's a cool example of
infotropism at the level of neurons, the retina in your eye,
(41:37):
the set of photoreceptors the back of your eye captures
the light of the world and tells the brain what
it's seeing out there. But it reads the world differently
in the day and in the night. So in bright
daylight there are plenty of photons to capture, and so
each photoreceptor just takes care of its own tiny dot
(41:58):
of the scene, and this gives nice and high resolution.
But at night it's a different story. There are few
photons to be captured, and so now the important job
of the retina is to detect that something was there,
even if it can't tell that with high resolution. So
at night, the photoreceptors do their work very differently. Essentially,
(42:21):
they join forces with their neighbors. They change the details
of their internal molecular cascades so they can link arms
with one another. Now they're operating as a team, they
can detect much lower levels of light, they can detect
fewer photons out there. So what we have here is
this beautiful and sophisticated strategy that has the retina operate
(42:46):
differently as the light levels go up or down. When
it's bright out, the system achieves high spatial resolution, and
when it's dark, photoreceptors pool together to have a better
chance to catch photons, which results in vision that's more
sensitive to dim light but blurrier in resolution. You've surely
(43:08):
noticed this at nighttime when you can't quite make out
what something is out there, whereas it would be easy
in the daytime. So the point here is that your eyes,
your retinas, put enormous work into shifting themselves around to
maximize information that matters whether the photons are plentiful or
they're rare. The retina optimizes itself to get useful data.
(43:33):
In the day, it captures the most detail so that
it can spot the rabbit at the distance. In dim light,
it shifts to higher sensitivity to capture whatever's out there
with lower detail, capturing the shadowy essence of the jaguar
lurking in the gloom. Mother Nature figured out not only
how to build an eye, but also how to adjust
(43:56):
its circuitry on the fly, so it can operate well
in different contexts, all to make the best use of
what's available. It is infotropic. Just like the plant seeks light,
the retina seeks data, and more generally, as we've seen
in this episode, the brain is always seeking information. It
(44:18):
constantly adjusts its circuitry to maximize the data it can
draw from the world. So if you're staring at a waterfall,
it discounts the downward motion so it can be sensitive
to other things. If you put on yellow sunglasses, your
brain discounts that shift in color so that it can
be more sensitive to other information. And you can detect
(44:41):
this shift when you take off the sunglasses and you
have a big blue color after effect. So to wrap up,
this episode plugs into the big picture we've been talking
about in many episodes, and we'll see in many more,
which is that the brain is locked in silence and darkness,
and its job is to build an internal model of
the outside world. And what we're seeing a little more
(45:04):
clearly now is that its internal model is really there
to make predictions. If the world proceeds as expected, the
brain saves energy. So when the world changes, let's say,
becoming a place with lots of downward motion or yellow
light or something, the brain adjusts to match it so
that it can be maximumly positioned to gather new information
(45:28):
that's unexpected. So these after effect illusions seem funny and
may be trivial, but they're actually a powerful way for
us to reveal this constant adaptation that your brain is doing.
Over the course of many episodes coming up, we're going
to see how fundamentally the brain is a prediction machine,
(45:48):
and that is the driving engine behind its constant self reconfiguration.
By modeling the state of the world, the brain reshapes
itself to have good expectation and therefore to be maximally
sensitive to the unexpected. Go to Eagleman dot com slash
(46:10):
podcast for more information and to find further reading. Send
me an email at podcasts at eagleman dot com with
questions or discussions, and I'll be making an episode soon
in which I address those. Until next time, I'm David Eagleman,
and this is Inner Cosmos.
Host
David Eagleman
Popular Podcasts
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com
24/7 News: The Latest
The latest news in 4 minutes updated every hour, every day.
Therapy Gecko
An unlicensed lizard psychologist travels the universe talking to strangers about absolutely nothing. TO CALL THE GECKO: follow me on https://www.twitch.tv/lyleforever to get a notification for when I am taking calls. I am usually live Mondays, Wednesdays, and Fridays but lately a lot of other times too. I am a gecko.